Files
cell/source/qjs_text.c
2025-11-25 21:05:48 -06:00

307 lines
10 KiB
C
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

#include "cell.h"
#include "blob.h"
#include <string.h>
#include <stdlib.h>
JSC_CCALL(text_blob_to_hex,
size_t blob_len;
void *blob_data = js_get_blob_data(js, &blob_len, argv[0]);
if (!blob_data) return JS_ThrowTypeError(js, "Expected stone blob");
uint8_t *bytes = (uint8_t *)blob_data;
size_t hex_len = blob_len * 2;
char *hex_str = malloc(hex_len + 1);
if (!hex_str) return JS_ThrowOutOfMemory(js);
static const char hex_digits[] = "0123456789ABCDEF";
for (size_t i = 0; i < blob_len; ++i) {
hex_str[i * 2] = hex_digits[(bytes[i] >> 4) & 0xF];
hex_str[i * 2 + 1] = hex_digits[bytes[i] & 0xF];
}
hex_str[hex_len] = '\0';
JSValue val = JS_NewString(js, hex_str);
free(hex_str);
return val;
)
JSC_CCALL(text_blob_to_base32,
size_t blob_len;
void *blob_data = js_get_blob_data(js, &blob_len, argv[0]);
if (!blob_data) return JS_ThrowTypeError(js, "Expected stone blob");
uint8_t *bytes = (uint8_t *)blob_data;
static const char b32_digits[] = "ABCDEFGHIJKLMNOPQRSTUVWXYZ234567";
// Calculate exact output length needed
size_t groups = (blob_len + 4) / 5; // Round up to next group of 5
size_t b32_len = groups * 8;
char *b32_str = malloc(b32_len + 1);
if (!b32_str) return JS_ThrowOutOfMemory(js);
size_t in_idx = 0;
size_t out_idx = 0;
while (in_idx < blob_len) {
// Read up to 5 bytes into a 40-bit buffer
uint64_t buf = 0;
int bytes_to_read = (blob_len - in_idx < 5) ? (blob_len - in_idx) : 5;
for (int i = 0; i < bytes_to_read; ++i) {
buf = (buf << 8) | bytes[in_idx++];
}
// Pad buffer to 40 bits if we read fewer than 5 bytes
buf <<= 8 * (5 - bytes_to_read);
// Extract 8 groups of 5 bits each
for (int i = 0; i < 8; ++i) {
b32_str[out_idx++] = b32_digits[(buf >> (35 - i * 5)) & 0x1F];
}
}
// Add padding if necessary
if (blob_len % 5 != 0) {
static const int pad_count[] = {0, 6, 4, 3, 1}; // padding for 0,1,2,3,4 bytes
int padding = pad_count[blob_len % 5];
for (int i = 0; i < padding; ++i) {
b32_str[b32_len - 1 - i] = '=';
}
}
b32_str[b32_len] = '\0';
JSValue val = JS_NewString(js, b32_str);
free(b32_str);
return val;
)
static int base32_char_to_val(char c) {
if (c >= 'A' && c <= 'Z') return c - 'A';
if (c >= 'a' && c <= 'z') return c - 'a';
if (c >= '2' && c <= '7') return c - '2' + 26;
return -1;
}
JSC_CCALL(text_base32_to_blob,
const char *str = JS_ToCString(js, argv[0]);
if (!str) return JS_ThrowTypeError(js, "Expected string");
size_t str_len = strlen(str);
if (str_len == 0) {
JS_FreeCString(js, str);
return JS_ThrowTypeError(js, "Empty base32 string");
}
// Remove padding to get effective length
size_t effective_len = str_len;
while (effective_len > 0 && str[effective_len - 1] == '=') {
effective_len--;
}
// Calculate output length: each group of 8 base32 chars -> 5 bytes
size_t output_len = (effective_len * 5) / 8;
uint8_t *output = malloc(output_len);
if (!output) {
JS_FreeCString(js, str);
return JS_ThrowOutOfMemory(js);
}
size_t in_idx = 0;
size_t out_idx = 0;
// Process in groups of 8 characters (40 bits -> 5 bytes)
while (in_idx < effective_len) {
uint64_t buf = 0;
int chars_to_read = (effective_len - in_idx < 8) ? (effective_len - in_idx) : 8;
// Read up to 8 base32 characters into buffer
for (int i = 0; i < chars_to_read; ++i) {
int val = base32_char_to_val(str[in_idx++]);
if (val < 0) {
free(output);
JS_FreeCString(js, str);
return JS_ThrowTypeError(js, "Invalid base32 character");
}
buf = (buf << 5) | val;
}
// Calculate how many bytes we can extract
int bytes_to_extract = (chars_to_read * 5) / 8;
// Shift buffer to align the most significant bits
buf <<= (40 - chars_to_read * 5);
// Extract bytes from most significant to least significant
for (int i = 0; i < bytes_to_extract && out_idx < output_len; ++i) {
output[out_idx++] = (buf >> (32 - i * 8)) & 0xFF;
}
}
JSValue val = js_new_blob_stoned_copy(js, output, output_len);
free(output);
JS_FreeCString(js, str);
return val;
)
static int base64_char_to_val_standard(char c) {
if (c >= 'A' && c <= 'Z') return c - 'A';
if (c >= 'a' && c <= 'z') return c - 'a' + 26;
if (c >= '0' && c <= '9') return c - '0' + 52;
if (c == '+') return 62;
if (c == '/') return 63;
return -1;
}
static int base64_char_to_val_url(char c) {
if (c >= 'A' && c <= 'Z') return c - 'A';
if (c >= 'a' && c <= 'z') return c - 'a' + 26;
if (c >= '0' && c <= '9') return c - '0' + 52;
if (c == '-') return 62;
if (c == '_') return 63;
return -1;
}
/*─── blob → Base64 (standard, with + and /, padded) ───────────────────*/
JSC_CCALL(text_blob_to_base64,
size_t blob_len;
void *blob_data = js_get_blob_data(js, &blob_len, argv[0]);
if (!blob_data) return JS_ThrowTypeError(js, "Expected stone blob");
const uint8_t *bytes = blob_data;
static const char b64[] = "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
"abcdefghijklmnopqrstuvwxyz"
"0123456789+/";
size_t out_len = ((blob_len + 2) / 3) * 4;
char *out = malloc(out_len + 1);
if (!out) return JS_ThrowOutOfMemory(js);
size_t in_i = 0, out_i = 0;
while (in_i < blob_len) {
uint32_t buf = 0;
int to_read = (blob_len - in_i < 3 ? blob_len - in_i : 3);
for (int j = 0; j < to_read; ++j) {
buf = (buf << 8) | bytes[in_i++];
}
buf <<= 8 * (3 - to_read);
out[out_i++] = b64[(buf >> 18) & 0x3F];
out[out_i++] = b64[(buf >> 12) & 0x3F];
out[out_i++] = (to_read > 1 ? b64[(buf >> 6) & 0x3F] : '=');
out[out_i++] = (to_read > 2 ? b64[ buf & 0x3F] : '=');
}
out[out_len] = '\0';
JSValue v = JS_NewString(js, out);
free(out);
return v;
)
/*─── Base64 → blob (standard, expects + and /, pads allowed) ────────────*/
JSC_CCALL(text_base64_to_blob,
const char *str = JS_ToCString(js, argv[0]);
if (!str) return JS_ThrowTypeError(js, "Expected string");
size_t len = strlen(str);
// strip padding for length calculation
size_t eff = len;
while (eff > 0 && str[eff-1] == '=') eff--;
size_t out_len = (eff * 6) / 8;
uint8_t *out = malloc(out_len);
if (!out) { JS_FreeCString(js, str); return JS_ThrowOutOfMemory(js); }
size_t in_i = 0, out_i = 0;
while (in_i < eff) {
uint32_t buf = 0;
int to_read = (eff - in_i < 4 ? eff - in_i : 4);
for (int j = 0; j < to_read; ++j) {
int v = base64_char_to_val_standard(str[in_i++]);
if (v < 0) { free(out); JS_FreeCString(js, str);
return JS_ThrowTypeError(js, "Invalid base64 character"); }
buf = (buf << 6) | v;
}
buf <<= 6 * (4 - to_read);
int bytes_out = (to_read * 6) / 8;
for (int j = 0; j < bytes_out && out_i < out_len; ++j) {
out[out_i++] = (buf >> (16 - 8*j)) & 0xFF;
}
}
JSValue v = js_new_blob_stoned_copy(js, out, out_len);
free(out);
JS_FreeCString(js, str);
return v;
)
/*─── blob → Base64URL (no padding, - and _) ─────────────────────────────*/
JSC_CCALL(text_blob_to_base64url,
size_t blob_len;
void *blob_data = js_get_blob_data(js, &blob_len, argv[0]);
if (!blob_data) return JS_ThrowTypeError(js, "Expected stone blob");
const uint8_t *bytes = blob_data;
static const char b64url[] = "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
"abcdefghijklmnopqrstuvwxyz"
"0123456789-_";
size_t raw_len = ((blob_len + 2) / 3) * 4;
// well drop any trailing '='
char *out = malloc(raw_len + 1);
if (!out) return JS_ThrowOutOfMemory(js);
size_t in_i = 0, out_i = 0;
while (in_i < blob_len) {
uint32_t buf = 0;
int to_read = (blob_len - in_i < 3 ? blob_len - in_i : 3);
for (int j = 0; j < to_read; ++j) {
buf = (buf << 8) | bytes[in_i++];
}
buf <<= 8 * (3 - to_read);
out[out_i++] = b64url[(buf >> 18) & 0x3F];
out[out_i++] = b64url[(buf >> 12) & 0x3F];
if (to_read > 1) out[out_i++] = b64url[(buf >> 6) & 0x3F];
if (to_read > 2) out[out_i++] = b64url[ buf & 0x3F];
}
out[out_i] = '\0';
JSValue v = JS_NewString(js, out);
free(out);
return v;
)
/*─── Base64URL → blob (accepts - / _, no padding needed) ─────────────────*/
JSC_CCALL(text_base64url_to_blob,
const char *str = JS_ToCString(js, argv[0]);
if (!str) return JS_ThrowTypeError(js, "Expected string");
size_t len = strlen(str);
size_t eff = len; // no '=' in URLsafe, but strip if present
while (eff > 0 && str[eff-1] == '=') eff--;
size_t out_len = (eff * 6) / 8;
uint8_t *out = malloc(out_len);
if (!out) { JS_FreeCString(js, str); return JS_ThrowOutOfMemory(js); }
size_t in_i = 0, out_i = 0;
while (in_i < eff) {
uint32_t buf = 0;
int to_read = (eff - in_i < 4 ? eff - in_i : 4);
for (int j = 0; j < to_read; ++j) {
int v = base64_char_to_val_url(str[in_i++]);
if (v < 0) { free(out); JS_FreeCString(js, str);
return JS_ThrowTypeError(js, "Invalid base64url character"); }
buf = (buf << 6) | v;
}
buf <<= 6 * (4 - to_read);
int bytes_out = (to_read * 6) / 8;
for (int j = 0; j < bytes_out && out_i < out_len; ++j) {
out[out_i++] = (buf >> (16 - 8*j)) & 0xFF;
}
}
JSValue v = js_new_blob_stoned_copy(js, out, out_len);
free(out);
JS_FreeCString(js, str);
return v;
)
static const JSCFunctionListEntry js_text_funcs[] = {
MIST_FUNC_DEF(text, blob_to_hex, 1),
MIST_FUNC_DEF(text, blob_to_base32, 1),
MIST_FUNC_DEF(text, base32_to_blob, 1),
MIST_FUNC_DEF(text, blob_to_base64, 1),
MIST_FUNC_DEF(text, base64_to_blob, 1),
MIST_FUNC_DEF(text, blob_to_base64url, 1),
MIST_FUNC_DEF(text, base64url_to_blob, 1),
};
JSValue js_text_use(JSContext *js)
{
JSValue mod = JS_NewObject(js);
JS_SetPropertyFunctionList(js, mod, js_text_funcs, countof(js_text_funcs));
return mod;
}